The System for Integrated Modeling of the Atmosphere (SIMA) project aims to unify existing NCAR community atmosphere modeling efforts across weather, climate, chemistry, and geospace research. NCAR scientists, in partnership with the atmospheric and geospace sciences research community, are developing a SIMA framework and infrastructure that enables simulations of atmospheric processes and atmospheric interactions with other components of the coupled Earth system ranging from the surface to the ionosphere, and across scales from cloud-resolving weather to decadal climate studies. One of SIMA’s goals is to enhance atmospheric and earth-system modeling applications for frontier-science problems. An example is the sub-seasonal to seasonal predictability of tropical cyclone formation, which requires the capability to represent convective-permitting scales over the tropics coupled to an Earth System Model (ESM). This can extend to investigating the role of aerosols on tropical cyclone formation. Another frontier-science application is quantifying the impact of biomass burning on air quality, atmospheric chemistry, and weather from local to global scales. This requires the capability to represent fires on convective-permitting scales and detailed chemistry and aerosol processes. Additional frontier science application examples are described in the SIMA Vision document. 

SIMA will allow NCAR to shift from using a complex modeling ecosystem composed of several atmosphere models, each with their own specific application (e.g., Weather Research and Forecasting (WRF) and Model for Prediction Across Scales (MPAS) standalone atmosphere models for weather research, Community Atmosphere Model (CAM) for climate research, Whole Atmosphere Community Climate Model eXtension (WACCM-X) for thermosphere and ionosphere research) into a single modeling system that can be configured for a range of applications (Figure 1). In November 2021, SIMA version 1 was released to the community. This initial version of SIMA includes development of a CAM configuration that contains regionally-refined grids over the Arctic and Greenland and development of high-resolution capability of WACCM-X and one-way coupling between WACCM-X and a geomagnetic grid mesh for magnetohydrodynamics calculations. For SIMA v1, atmospheric chemistry input, emissions data, and chemistry code have been modified to be compatible with unstructured grid meshes and regional refinement of grids. Atmospheric chemistry simulation output for a CAM configuration with regional refinement over the contiguous US was made available via the Geoscience Data Exchange site. A model-independent chemistry module, which enhances the flexibility of prescribing the chemical constituents and reactions, was released in a box model configuration for testing and classroom teaching.

When SIMA is mature, it will provide functionality for interoperable model components, including physics and chemistry schemes, as well as dynamical cores, by using the Community Common Physics Package (CCPP) in its functionality. A suite of physics parameterizations from WRF and CAM are being modified to be CCPP compliant and thus, will become available as part of SIMA. Recently, software engineers and scientists at NCAR, the DTC, and NOAA held a series of discussions about the usefulness of CCPP for the physics suite in CAM as well as any future SIMA needs that CCPP does not yet address. In summary, the CCPP should improve the efficiency of future CAM code development and maintenance, which will enable other modeling system advancements. CCPP is easy to build upon, is explicit, and its implementation facilitates the detection of bugs and inconsistencies. At the same time, it is understood that the CCPP framework will not do everything needed or desired for the modeling system, and additional steps will need to be taken to achieve specific requirements.

One major achievement in developing SIMA is implementing the MPAS dynamical core in CAM giving CAM new functionality to resolve convective motions with a non-hydrostatic dynamical core. Several tests are being conducted using a global MPAS mesh at 60-km grid spacing with regional-refinement to 3-km grid spacing over a specified region. An example application from the NSF-funded EarthWorks project demonstrates the capability of predicting precipitation amounts over the Pacific Northwest region of the US (Figure 2). 

During the past year, the SIMA governance structure has been broadened to establish a SIMA steering committee, a SIMA Project Lead, a SIMA Scientific and Technical Co-Leads group, and a SIMA external advisory panel. Under this expanded structure, SIMA is taking a two-pronged approach to continued infrastructure development. The first is continuing to produce capabilities already identified as important. For example, there is planned work for enabling simulations with the SIMA-provided configuration of using the non-hydrostatic MPAS dynamical core in CAM using a global grid spacing of 3.75-km, which would provide the ability to perform subseasonal to seasonal forecasts in an Earth System Model. Other high-priority targets include refactoring CAM physics to be compliant with CCPP; developing an online, flexible regridding tool to improve input preprocessing for desired model grids; and enhancing the Model Independent Chemistry Module, while implementing it into CAM.

Another path for SIMA development is to identify and pursue a frontier-science application that will guide infrastructure development. NCAR has asked their staff to propose projects that require SIMA to develop additional functionality that can be utilized in investigations that will understand processes or predictability from local to regional to global scales and synthesize cross-disciplinary science. The SIMA developments for this science application should open the door for many other groups to apply SIMA for advancing their own science interests. 

SIMA leadership is looking forward to deeper engagement with the community to further advance the single atmospheric modeling system and conduct exciting, new science. See more information about SIMA at sima.ucar.edu.

The DTC was established in 2003 as a multi-agency effort with funding from National Oceanic and Atmospheric Administration (NOAA), the US Air Force, and the National Center For Atmospheric Research (NCAR) and has made its mark as the “clearing-house” for testing and evaluation (T&E) activities within the meteorological and associated Earth science community. As such, it provides a fundamental bridge between research and operations where cutting-edge ideas can be explored and vetted to enhance understanding of NWP physical processes and improve forecast verification with various techniques and metrics. The DTC's mission priorities overlap with its fellow NOAA testbeds, as they are unique collaborative spaces where researchers and forecasters work together to improve weather products and services (https://wpo.noaa.gov/testbeds/). Thus, the DTC’s role and involvement in partner NOAA testbeds have broadened for more than a decade.

The DTC’s participation in NOAA’s Hazardous Weather Testbed (HWT) Spring Forecasting Experiments (SFE) happened to coincide with my role as a facilitator of the SFE in 2010 and 2011 while working at the NOAA/NWS Storm Prediction Center (SPC). The DTC’s role was to conduct objective verification of the experimental model forecasts and provide results as feedback to the participants in the evaluation process (Clark et al. 2012; https://doi.org/10.1175/BAMS-D-11-00040.1). For this evaluation, DTC applied its locally developed Model Evaluation Tools (MET). Traditionally, the SFE examines experimental high-resolution (convection-allowing) models and ensembles that can explicitly simulate convective mode, and provide details about potential hazards (tornadoes, severe hail, or strong-straight line winds). While the storm-attribute fields often appear realistic, conventional grid-point verification methods routinely penalize high-resolution forecasts when small offsets in time and space exist between observed and forecast event objects. As a result, more emphasis has been placed on spatial techniques that avoid the inherent “double penalty problem” (i.e., the standard verification metrics produce both a miss and false alarm for what subjectively appears a reasonable forecast). Under this framework, a more reliable practice of object-oriented verification was applied by the DTC by applying MET’s Method for Object-based Diagnostic Evaluation (MODE). Some results from the 2010 SFE using MODE are available in Clark et al. (2012). Alternatively, starting with the 2012 SFE, neighborhood-type evaluations also tended to be a popular choice with the participants, as the statistics (e.g., Fractions Skill Score) were made available in near real-time the next day, along with spatial plots on dedicated webpages (Melick et al. 2012; https://www.spc.noaa.gov/publications/melick/hwt2012.pdf).

MET was developed over several years, continually maintained, and expanded by DTC, and has served the academic, government, and private sectors, as well as the international community. DTC provides a consistent code repository for MET with robust documentation, training, and support, which are always improving. These capabilities and resources are essential for sustaining both research and operational purposes. In 2017, I made the transition to becoming a meteorologist for the 16th Weather Squadron (within the 557th Weather Wing) at Offutt Air Force Base, Nebraska. An extensive collaborative history exists between the Air Force and DTC as they’ve funded the DTC on specific T&E projects related to improving understanding weather phenomena and ultimately, environmental intelligence on a global level. Some of the noteworthy byproducts from these investigations have been incorporated into the MET software upgrades. In the past, MODE had been used by the 16th Weather Squadron for case-study evaluations, although not on a routine basis. The adoption of the remainder of the MET tools into an operational context for objective verification has been gradual over the last five years. Our experience with DTC has been very encouraging as there is often a quick turnaround when troubleshooting problems with the software suite and providing solutions (as well as identifying shortcomings that are often addressed in subsequent releases).

The proper infusion of science into T&E activities at DTC has been guided on an annual basis through review, evaluation, and recommendations from the DTC Science Advisory Board (SAB). The SAB convenes experts from all fields in the society that can help to shape the strategic direction and objectives for DTC. For my second year on the DTC SAB, I was honored to serve as one of the chairs during the Fall of 2022. During the three days of meetings in September and October, my responsibilities included guiding SAB discussions and giving a presentation on my career insights on the DTC, which included offering suggestions on limitations in the field that could be addressed in the future. Finally, with feedback from other members on the board, I oversaw the development of our final report delivering constructive critique and advice to DTC. I am extremely grateful on both a personal and professional level to have this experience and I view it as rewarding to the US Air Force to serve on their behalf and promote their interests and goals, with respect to weather challenges.

Molly Smith is one of DTC’s NOAA collaborators on our staff, and recently became one of the leads for METplus. Her background is in tropical meteorology, and she is the primary developer for the METexpress visualization system.

Molly grew up in the chaparral on the edge of the Mojave Desert in Southern California. It does not rain very often in that part of the country, but when Molly was five, a strong El Niño flooded her hometown, turning the streets into rivers. From that moment on, she became very interested in learning how the weather works. Her parents, who are both artists, were very surprised that they had somehow produced a meteorologist, but encouraged her by getting lots of books about the weather and later sending her to science camp. Molly went on to get a bachelor’s in atmospheric science at Cornell University, and then a master’s in the same field at the University at Albany, where she examined various models’ performances in simulating the landfall of Hurricane Irene (2011).

While at the University at Albany, Molly realized that she really enjoyed writing software, and that her ideal career would be one that combined both meteorology and programming. She was able to find exactly what she was looking for, and in 2017 started work as part of the NOAA Global Systems Laboratory (GSL) model verification team. There, she became the lead developer for GSL’s Model Analysis Tool Suite (MATS), which continues to be one of her primary responsibilities to this day. MATS is a lightweight, quick-start visualization suite which allows GSL’s model developers to quickly compare the skill of individual models and ensembles, as well as assess the performance of models under development. MATS also enables developers to produce publication-quality interactive graphs in a web-based framework, without needing to write their own code, and to perform a focused analysis of interesting weather features.

In 2019, the DTC began work on a spin-off of MATS that would display statistics produced by its METplus verification suite, with Molly again as the lead developer. This project became known as METexpress. METexpress was intended to work in tandem with the existing METplus visualization suite, METviewer. METviewer is a very powerful tool for performing in-depth analyses of verification data, but it can have a learning curve for new users. METexpress, therefore, is intended to fill this gap by giving users an intuitive interface with which they can easily produce common plot types and assessments. METexpress has greatly aided the adoption of METplus at GSL, allowing model developers to produce unified verification system-derived statistical plots in a familiar setting. This year Molly became the GSL team lead for METplus, and hopes to contribute further to the evolution of this excellent and versatile verification software package.

Outside of work, Molly can usually be found going for long walks, hiking, and catering to three very opinionated cats. She also has an extensive book collection and will eventually need to find places for more shelving. She regards this as a good problem to have.

A constant struggle in NWP model design is the tradeoff between scientific improvements and computational cost. A method commonly used to balance that tradeoff is lowering some numerical calculations to single precision (or 32-bit calculations). Often, single-precision calculations have enough precision for physics, and they reduce disk storage, memory usage, and computation time. To apply single-precision calculations correctly, it is necessary to carefully evaluate and fine-tune algorithms, and perhaps in isolated places of the code, perform calculations in double precision (64 bits).

This approach is already used in several operational models. For example, the NOAA operational RAP and HRRR models (which use the WRF model) primarily use single-precision physics and the ECMWF Integrated Forecasting System (IFS) model recently switched the bulk of its physics calculations from double to single precision. However, until recently, the DTC-hosted Common Community Physics Package (CCPP), which is directly used by the UFS modeling system, was missing this key capability.

Earlier this year, the US Naval Research Laboratory (NRL) successfully enabled a single-precision physics suite in the Navy Environmental Prediction System Using the NUMA CorE (NEPTUNE) model. They focused on developing the RAP software suite available via the CCPP, leveraging previous work done over the years by the broad community that developed the RAP model. This suite had previously been widely tested and carefully tuned to work well with most of the code in single precision, while retaining key parts of the code in double precision. NRL’s work corrected some imprecise calculations and troublesome constants, addressing issues that were not present in the WRF version of the same parameterizations. 

This technical achievement by NRL, which was made available to the CCPP authoritative code repository, paved the way for the DTC to make the single-precision RAP suite more widely available. The DTC generalized the code and added support to the UFS, which employs the FV3 dynamical core. Since the UFS already supported both single and double precision for dynamics, the logic of mixed precision already existed, at least conceptually. Therefore, most of the work was of a mechanical nature: fixing type mismatches in the code and in connections to coupling, libraries, and stochastic physics.

The outcome of this work is that the single-precision RAP suite now works technically in the UFS. So far it has only been tested at low resolution, and results indicate that it runs approximately 25% faster than at double precision. 

There is much work to be done before single-precision physics in CCPP can be considered a finished product. First, scientific validation of the RAP suite has to be conducted, which may reveal the need for additional code adjustments. Second, further development could reduce computation cost of calculations. UFS only reads and writes files with double-precision floating point, and some areas of the code may still convert between single and double precision unnecessarily. Finally, additional suites could be made to work in single precision. This work has laid the foundation for budding capability on which the DTC and the community will be able to build, but only if agencies continue to invest in this development.

Continuous integration (CI) is a software development practice where developers regularly merge their code changes into a central repository whereby automated builds and tests are run. The key goals of continuous integration are to quickly find and address bugs, improve software quality, and reduce the time it takes to validate and release new software updates. The DTC Common Community Physics Package (CCPP) team has not only adopted CI for CCPP Physics code updates and releases, but also applied the CI concept to community engagement and support. The CCPP team established bi-weekly CCPP Physics code management meetings in August 2021. Representatives of organizations, NOAA, NCAR, Naval Research Laboratory (NRL), and DTC, regularly involved in the development of  CCPP actively participate in the meetings to discuss issues, develop solutions, and evaluate outcomes in a productive collaborative forum.

One successful example of the CI for community engagement in CCPP Physics code management is the establishment of a code fork referred to as the UFS Fork of CCPP Physics. As more groups started using and developing CCPP-compliant physics and chemistry, an effective community collaboration methodology on CCPP Physics has become increasingly important. After collecting community input in late 2020, the DTC CCPP team developed a proposal for the CCPP Physics repository structure and code management plan in June 2021. The proposal includes a code repository structure consisting of the authoritative CCPP Physics main branch maintained by the CCPP team and multiple CCPP Physics forks for different model systems maintained by corresponding modeling centers or developers. For the individual physics scheme, a similar structure will be adopted, in which individual physics scheme developers are responsible for authoritative physics repositories, and modeling centers are responsible for their fork/branches. The proposal has been discussed and improved iteratively at the CCPP Physics code management meetings for a year and the proposed recommendations were implemented in October 2022. The DTC CCPP team and EMC are currently co-managing the UFS Fork of CCPP Physics.

CI for community input and engagement in CCPP Physics code management has led to many important outcomes and initiatives in addition to the UFS Fork of CCPP Physics. The engagement of the NCAR Mesoscale & Microscale Meteorology (MMM) model development team led to a new DTC project to test the CCPP compliance of NCAR MMM physics suite in the CCPP Single Column Model (SCM). The NRL team identified a backward incompatibility issue for their Navy Environmental Prediction sysTem Utilizing the NUMA corE (NEPTUNE) model when CCPP Physics updates required corresponding changes in the host model this summer. The DTC CCPP team actively worked with the NRL team to propose a solution for this problem using CCPP Framework functionality and the CI capability offered by GitHub. NCEP EMC representatives recently raised the question of how to simplify and improve the CCPP interstitial schemes (​​modularized pieces of code that perform data preparation, diagnostics, or other "glue" functions, and allow primary schemes to work together as a suite), which was echoed among all groups. The DTC CCPP team built an inventory of all existing interstitial schemes and started to classify them based on their host model specificity. Proposed pathways addressing this problem were discussed and refined at multiple meetings. This ongoing effort is expected to lead to improvement in the interoperability of CCPP Physics.

The concept of CI is useful in software development and is very helpful in connecting communities. The DTC CCPP team will continue to engage the broader community through different platforms and opportunities. The planned CCPP visioning workshop in 2023 will include a topic on how to better engage and support the community in using and developing CCPP. 

NOAA is offering paid summer internships, named the 2023 Lapenta Internship, targeted towards current 2nd and 3rd-year undergraduate and enrolled graduate students to work in areas that will provide robust research and/or operational experience to prepare them for further study in NOAA fields, for application to fellowships, or for the NOAA-mission workforce.

Projects may be focused on research areas or the development of operational products such as decision support tools, climate and weather forecast models, population dynamics of fish populations, ecosystem modeling, hydrology, ocean circulation models, unmanned systems (both in air and underwater), data analysis methodologies, social science, and strategies to communicate climate and weather information pertaining to NOAA's mission to the public and to stakeholders. Overall, students will focus on areas that will meet the future needs of NOAA’s ever-broadening user community and address immediate challenges across a diverse spectrum from the sun to the bottom of our oceans. This internship program enables the many NOAA agencies to target the skills needed to fulfill their specific mission needs.

The application period for the 2023 Lapenta Internship will be open from Oct 1, 2022 to Jan 5, 2023 (about one extra week will be allowed for processing of recommendation letters). The 2023 Lapenta Internship will run from June 5, 2023 to August 11, 2023.

Eligibility information: 

• Must be enrolled in an undergraduate (sophomore and junior status only) or graduate degree program, and be a U.S. Citizen willing to undergo a security background check. US citizenship must be in hand by Jan 1, 2023.
• Students from all STEM majors relevant to NOAA's mission will be considered.
• NOAA’s Diversity and Inclusion Strategic Plan applies, where members of groups underrepresented in the sciences are strongly encouraged to apply.

For more information, visit the 2023 Lapenta internship website.

UCAR’s Cooperative Programs for the Advancement of Earth System Science (CPAESS) is excited to announce the launch of a new fellowship with NOAA's Weather Program Office (WPO): the WPO Innovation for Next Generation Scientists (WINGS) Dissertation Fellowship. The WINGS Fellowship is designed for Ph.D. candidates who have completed their required coursework, and are in the beginning stages of writing a dissertation. Fellows will work with their academic advisor and a mentor recommended by WPO and CPAESS. This inaugural award year will focus on research relevant to NOAA's Earth Prediction Innovation Center (EPIC) Program mission. The Application Deadline is January 27, 2023.

Potential topics for the fellowship this year can focus on one or more areas of scientific and/or technical importance shared by UCAR and EPIC, including:   

• Data Assimilation
• Atmospheric Physics
• Systems Architecture (coupling, workflow, continuous integration and development)
• Machine Learning/Artificial Intelligence
• Software Engineering to advance numerical weather prediction

For more information, visit the Weather Program Office (WPO) Innovation for Next Generation Scientists (WINGS) Dissertation Fellowship webpage.